Answers are grounded in this paper and cite their sources.
Research explores binary segmentation methods on T1-weighted MRI, revealing implications for efficient brain tumor classification.
Brain tumors consist of abnormal tissues resulting from uncontrolled cell proliferation and have no physiological function in the brain. The application of binary image segmentation is important because it is widely used in medical imaging to assist in brain tumor diagnosis. There has been prior work on MRI image classification for brain tumor types, such as using EfficientNet. This study aims to investigate the application of binary segmentation results from T1-weighted contrast MRI images in the EfficientNet model. The study employed a Research and Development (R&D) methodology. The dataset comprised 1,400 images, including 700 glioma and 700 meningioma cases. MRI images and tumor masks were combined using a binary masking segmentation method. Subsequently, the segmented images were input into the EfficientNet model and evaluated using a confusion matrix. The model testing yielded an accuracy of 0.6, precision of 0.627, recall of 0.492, and an F1 score of 0.551. However, the loss value was relatively high at 0.678. These results suggest that binary segmentation images from T1-weighted contrast MRI scans of brain tumor cases can be applied effectively to the EfficientNet model.
No takes yet. Share an insight, caveat, or question.
Ihtisyamuddin et al. (2025) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: